Grounding an inverter

[CharlieJ]

BTW, there is a very significant difference between bonding and grounding.

That said, I have quoted the NEC, ISO 13297 and E-11 and provided significant references for further education about the danger associated with not maintaining the "holy grail".

An AC source aboard a vessel that does not have a N-G bond is unsafe.

[sailorchic34]

I understand about the various shock paths and the danger in marina's. My issue or question is at anchor (not on shore power) with a transformer type inverter, there isn't shock path to ground (earth) or water but only if someone comes in contact to both legs.

There isn't a current path from one leg to ground? At least to my eyes. On shore power there very well can be, but at anchor??? Where do you see the circuit path with a transformer inverter.

BTW I have read the NEC, well parts of it....

[sailorchic34]

Originally Posted by CharlieJ

An AC source aboard a vessel that does not have a N-G bond is unsafe.

Please explain how a transformer type inverter (not an auto transformer) is unsafe with out a ground.

EDIT: for example, say touching either leg with one hand and the engine in the other with an ungrounded transformer inverter. Like I have on my boat.. There isn't a ground reference at anchor as current flow is ONLY from leg to leg and NOT leg to ground.

[Thumbs Up]

I know the difference between ac ground and bonding but on AYBC boats they are connected right? So if for some reason there is power leaking into the ac ground it can leak into the water around the boat. I do not see how this wire, which can potentially connect live ac to the water could make the water any safer than if the ac ground had been isolated. They say that the purpose of the green wire is to provide an alternate path to ground IN ORDER TO TRIP THE BREAKER, right? Isn't it likely at some point that a ground fault like this would just be partial or intermediate, not enough to trip a breaker but enough to energize the water or electrocute someone? And if you were using the proper gfci (or RCD or whatever you want to call them) that you wouldn't need this green wire protection in the first place. I am no electrical engineer but have a pretty good understanding of ac electrical systems. I do not think that "no swimming" signs in marinas are the answer. GFCI (RCD) is definitely the answer. But if we have those, is the green wire necessary? (anybody?) and is it necessary to connect that wire to the water? I don't think that the current codes say anything to protect against electric shock drowning (except for the sign) so I am not going to take "because the code says so" for an answer.

[sailorchic34]

There as so many ground paths with ac (shore power system) much of it happens because of faulty wiring on the dock (it happens), the boat or adjacent boat, switching power supplies, etc. It's a mess even with proper grounding.

YES GFCI is by far the safest way to de energizes circuits. Also making sure all the boat's wiring in correct. Alas many times its a adjacent boat with a hot neutral (reversed wiring, UGH!) causing ground loops. All too common.

BTW I use my inverter as an isolated power supply, Not grounded to AC. It does NOT connect to the AC panel. If you have a inverter connected with a transfer switch to your AC panel then Grounding is not optional.

Per code it's never optional. But having a fair understanding of how the electrons flow (very slooooowwwwly, weird) and having read the code and knowing how my inverter is made, I went sans-ground. YMMV

[DotDun]

Originally Posted by sailorchic34

AH but the inverters we are talking about don't generally have autotransformers. Nor are most residential 240v pole mount either. Lets not muddy the waters. For this discussion we're talking about magnetic coupling but not electrically coupled.

Not trying to muddy the waters. You made the statement that all transformers provide isolation, you didn't qualify to only transformers used in inverters. Since auto transformers are a viable appliance for a boat for step up/down voltage conversion, I thought is appropriate to clarify.

Originally Posted by sailorchic34

....... If you have a inverter connected with a transfer switch to your AC panel then Grounding is not optional.

Per code it's never optional. But having a fair understanding of how the electrons flow (very slooooowwwwly, weird) and having read the code and knowing how my inverter is made, I went sans-ground. YMMV

My inverter (w/internal xfer switch) is behind my isolation transformer, so, yes, grounding is optional, even when connected to shorepower. Albeit, against code. FWIW, my generator is connected to the primary side of my isolation transformer (via shorepower/off/generator switch) for the same reason.

[sailorchic34]

Originally Posted by DotDun

My inverter (w/internal xfer switch) is behind my isolation transformer, so, yes, grounding is optional, even when connected to shorepower. Albeit, against code. FWIW, my generator is connected to the primary side of my isolation transformer (via shorepower/off/generator switch) for the same reason.

I agree with this... I know weird right.

Few boats in my area have an isolation transformer. It is the best way to be safe.

Yep I did say all transformers, so thanks for bringing me down to earth. Pun intended...

[Frankly]

Concerning the OP I am/ was bothered when a mfg supplies a device with a 3 wire receptacle and then leaves the ground pin unconnected. Kind of like telling someone that a loaded gun is unloaded.

As to why ungrounded electrical systems are not more common, when done properly they are more costly and more complex. I used to design systems for Navy Divers to safely use electrical equipment underwater. Not easy or cheap but is done quite regularly. The safe AC current level at 50/60 Hz is normally accepted lower than 30 ma ( causing ventricular fibrillation). Let go currents are less than this but not all human electrical contact is of this type. 5 ma is normally considered a safe current level unless the path includes human internal organs (think operating room).

All real world electrical systems have leakage resistance and more importantly for AC systems capacitance. If one contacts a high side of ungrounded system and the impedance (resistance/ capacitance) + body resistance is such that the current flow is less than 5 ma then no harm (there will always be some current flow). If the conditions are such that the current flow is say 50 ma then the person may become the “canary in the mine”. “Houston, we have a problem”

Years back others decided that to connect one hot conductor to a third ground conductor and provide a 3 wire distribution system was a reasonable approach in the safety/ utility tradeoff. In NEC terminology this is called the “grounded conductor” or neutral (white or grey insulation). The “ground conductor” (note ground vice grounded) is the green conductor and normally only conducts electricity under electrical failure conditions (wire or electrical appliance). The breaker opens removing the power from the circuit/ appliance and unless you happen to have one hand on the toaster and one hand on the water faucet when the problem develops all is well. GFCI receptacles/ breakers, double insulated tools are all extensions/ improvements on this concept.

For this concept to work correctly there must be a large current flow to cause the breaker to open immediately to interrupt the electricity to the fault. On a standard shoreside power system this is usually not a problem. When the source is a solid state inverter with built in fast acting overcurrent/ fault sensor it is not clear that things will work out as intended. My guess is the ABYC folks looked at this and decided that connect the neutral and ground was the best choice but nearly as clear cut as the dirt dwelling situation. When the electrical feed is coming down the dock then this neutral/ ground connection is shifted back to the shore side service. ELCI pedestal breakers are just 30 ma GFCI breakers. For a single circuit it is not difficult to keep the leakage current below 5 ma but when the feed is multiple circuits the trip threshold needs to be raised. Remember, where there is voltage there is always current (leakage and/or intended) flowing.

[BlackHeron]

Originally Posted by sailorchic34

There as so many ground paths with ac (shore power system) much of it happens because of faulty wiring on the dock (it happens), the boat or adjacent boat, switching power supplies, etc. It's a mess even with proper grounding.

YES GFCI is by far the safest way to de energizes circuits. Also making sure all the boat's wiring in correct. Alas many times its a adjacent boat with a hot neutral (reversed wiring, UGH!) causing ground loops. All too common.

BTW I use my inverter as an isolated power supply, Not grounded to AC. It does NOT connect to the AC panel. If you have a inverter connected with a transfer switch to your AC panel then Grounding is not optional.

Per code it's never optional. But having a fair understanding of how the electrons flow (very slooooowwwwly, weird) and having read the code and knowing how my inverter is made, I went sans-ground. YMMV

I did something similar. My inverter is also not bonded, although I intercept the wiring for my GP receptacle circuit and use a 3-pole transfer switch to re-route all 3 wires including the ground of that circuit to the inverter and completely disconnect them from the panel.

The GP receptacle circuit goes from the transfer switch directly to a GFCI receptacle and all the other (3) receptacles on the boat are fed downstream of it.

The only time I ever need to use the inverter is when I'm not on shore power. When I'm on shore power I disconnect the inverter and use said shore power to energize my GP outlets. Nothing but cord-connected appliances are ever hooked up to the GP recep circuit. The inverter has its own primary (12v) disconnect battery switch and never even gets turned on unless we are out to sea, and then only when 120v is needed (rarely.)

Bascially I'm just using the GP recep circuit as an extension of the outlet on the inverter. No ships are given to the haters who think this is dangerous. They don't understand the theory. There is no way that anyone could be hurt by this system.

[ottow]

Originally Posted by Frankly

Concerning the OP I am/ was bothered when a mfg supplies a device with a 3 wire receptacle and then leaves the ground pin unconnected. Kind of like telling someone that a loaded gun is unloaded.

As to why ungrounded electrical systems are not more common, when done properly they are more costly and more complex. I used to design systems for Navy Divers to safely use electrical equipment underwater. Not easy or cheap but is done quite regularly. The safe AC current level at 50/60 Hz is normally accepted lower than 30 ma ( causing ventricular fibrillation). Let go currents are less than this but not all human electrical contact is of this type. 5 ma is normally considered a safe current level unless the path includes human internal organs (think operating room).

All real world electrical systems have leakage resistance and more importantly for AC systems capacitance. If one contacts a high side of ungrounded system and the impedance (resistance/ capacitance) + body resistance is such that the current flow is less than 5 ma then no harm (there will always be some current flow). If the conditions are such that the current flow is say 50 ma then the person may become the “canary in the mine”. “Houston, we have a problem”

Years back others decided that to connect one hot conductor to a third ground conductor and provide a 3 wire distribution system was a reasonable approach in the safety/ utility tradeoff. In NEC terminology this is called the “grounded conductor” or neutral (white or grey insulation). The “ground conductor” (note ground vice grounded) is the green conductor and normally only conducts electricity under electrical failure conditions (wire or electrical appliance). The breaker opens removing the power from the circuit/ appliance and unless you happen to have one hand on the toaster and one hand on the water faucet when the problem develops all is well. GFCI receptacles/ breakers, double insulated tools are all extensions/ improvements on this concept.

For this concept to work correctly there must be a large current flow to cause the breaker to open immediately to interrupt the electricity to the fault. On a standard shoreside power system this is usually not a problem. When the source is a solid state inverter with built in fast acting overcurrent/ fault sensor it is not clear that things will work out as intended. My guess is the ABYC folks looked at this and decided that connect the neutral and ground was the best choice but nearly as clear cut as the dirt dwelling situation. When the electrical feed is coming down the dock then this neutral/ ground connection is shifted back to the shore side service. ELCI pedestal breakers are just 30 ma GFCI breakers. For a single circuit it is not difficult to keep the leakage current below 5 ma but when the feed is multiple circuits the trip threshold needs to be raised. Remember, where there is voltage there is always current (leakage and/or intended) flowing.

An inverter that is double-isolated such as mine should have a 2-pole outlet intended for CEE 7/16 connectors according to local regulations. A CEE 7/4 (which is installed on my unit) is illegal in this application as the ground pin is not connected.
Not that such things ever bothered Chinese manufacturers...

[Ahmet Erkan]

Originally Posted by ottow

So, I have a eBay inverter (https://www.ebay.com/itm/Power-Inver...QAAOSwAF5UYusW) and while it works just fine for most use it does generate a slight bit of 50hz hum when used with anything audio-related.
The unit has no external ground/earth post, and came with no instructions re grounding.
Opening it up, I am finding that the ground connector in the outlet is not connected at all...
My multimeter detects no connection between the 12v posts and any of the outlet poles, nor the chassis.

I'd like to get rid of the hum if possible, but am not sure how.
Is it even possible?

Thanks,

Otto

I would try jumpering the chassis (heatsink) of the inverter to the chassis of the audio equipment. If this works you could make the connection inside the inverter. If no joy, jumper both chassis to the battery negative. As a last resort, jumper the output neutral of the inverter to the chassis of the inverter. Still no significant reduction in hum, buy a hum filter and jumper the filter's case to the chassis of the inverter as well.
Lastly, if none of the above works, buy a Spanish guitar and take lessons :-)
Good luck Otto.
Ahmet

[transmitterdan]

It's a bad idea to jumper the neutral of some inverters to the chassis and the DC minus bus. In general, it's a bad idea to change anything inside an inverter or any electronic device without full knowledge of the schematic and operating principles of the device.

[Zil]

Give that inverter to an unsuspecting local, and get yourself a quality inverter. One brand is Sampex. There are others. Then, no hum.